clutch hitting?
Steve Jaros
the notion of clutch hitting. How can anyone not after watch Jeter the past
week club 'clutch' hits vs. boston in game 7 and last night vs. the marlins?
The guy is an all-time great, any way you slice it...
--
"if federal judges have the final word over its meaning,
the Constitution would be a mere thing of wax in the hands
of the judiciary, which they may twist and shape into any form
they please".
- Thomas Jefferson
SavoyBG
>It's hard to believe that there are still some people who don't 'believe' in
>the notion of clutch hitting. How can anyone not after watch Jeter the past
>week club 'clutch' hits vs. boston in game 7 and last night vs. the marlins?
Maybe because he was 4 for 27 in the 2001 world series?
Bruce Grossberg
igor eduardo küpfer
<TrojanFamily...@trojannutcase.com> wrote in
<KRDlb.87293$sp2.75682@lakeread04>:
>It's hard to believe that there are still some people who don't 'believe' in
>the notion of clutch hitting.
*sigh* It's not a matter of belief -- it's a matter of evidence.
> How can anyone not after watch Jeter the past
>week club 'clutch' hits vs. boston in game 7 and last night vs. the marlins?
>
Nobody says clutch hits don't exist. The existence of clutch hitters is what's
being debated.
<...>
--
--------------------------------
| best, | Sticking it to |
| ed | The Man since 1971 |
--------------------------------
Watch the spam trap -- the domain is rogers
The Only Mosey
and remarked:
>
>It's hard to believe that there are still some people who don't 'believe' in
>the notion of clutch hitting. How can anyone not after watch Jeter the past
>week club 'clutch' hits vs. boston in game 7 and last night vs. the marlins?
Wow, Jeter is so good he can move Derrek Lee off the line, or hit it just where
he wants, but, um, not always, only when he tries really hard.
>The guy is an all-time great, any way you slice it...
I slice it lots of ways and he is a bit better than average.
John Mosey "Humping your theoretical mom since 1993"
Complaints can be sent to ab...@mosey.com
The sun rises in the east, dumbass
Ima Pseudonym
>It's hard to believe that there are still some people who don't 'believe' in
>the notion of clutch hitting. How can anyone not after watch Jeter the past
>week club 'clutch' hits vs. boston in game 7 and last night vs. the marlins?
>
>The guy is an all-time great, any way you slice it...
It was really clutch the way he just hit into a double play, while
jogging down the line instead of running.
M. Zaiem Beg
->Nobody says clutch hits don't exist. The existence of clutch hitters is what's
->being debated.
Or more specifically, the existence of clutch ability.
--
M. Zaiem Beg zb...@iglou.com
The oh so very proud sponsor of the Pete Rose page at
baseball-reference.com
Dale Hicks
> On Wed, 22 Oct 2003 17:50:35 -0500 Steve Jaros opened up that worthless piehole
> and remarked:
> >
> >It's hard to believe that there are still some people who don't 'believe' in
> >the notion of clutch hitting. How can anyone not after watch Jeter the past
> >week club 'clutch' hits vs. boston in game 7 and last night vs. the marlins?
>
> Wow, Jeter is so good he can move Derrek Lee off the line,
Lee was playing off the line? I didn't notice. Seems like something the
announcers should have mentioned.
--
Cranial Crusader dgh 1138 at bell south point net
Steve Cutchen
<TrojanFamily...@trojannutcase.com> wrote:
> It's hard to believe that there are still some people who don't 'believe' in
> the notion of clutch hitting. How can anyone not after watch Jeter the past
> week club 'clutch' hits vs. boston in game 7 and last night vs. the marlins?
He's not a clutch hitter.
He was just due.
Steve Cutchen
<dgh...@yahoo.com> wrote:
HIS FATHER IS THE DISTRICT ATTORNEY!!
Mpoconnor7
>hat's
>->being debated.
>
>Or more specifically, the existence of clutch ability.
Some people consider Reggie Jackson to be one of the greatest clutch hitters of
all time primarily on the basis of a single WS game when he hit three home
runs. According to baseball-reference.com, Reggie hit .357 with 10 HR's and
24 RBI in 27 WS games which are outstanding numbers. If he was really such a
great clutch hitter, why did he only hit .238 with 8 HR's and 24 RBI in 50 LCS
games?
In a total of 77 postseason games, Mr. October hit .278 with 18 HR and 48 HR in
a 77 games; roughly half a season of LCS and WS games. If you double those
totals to approximate a 162 game season, you have .278 with 36 HR and 96 RBI in
154 games; his career average per 162 games is .262 BA with 32 HR and 98 RBI.
In other words his PS performance compares favorably to his regular season
stats although the BA is a little bit higher.
I'm not really convinced that Reggie was such a great clutch hitter.
Michael O'Connor - Modern Renaissance Man
"The likelihood of one individual being correct increases in a direct
proportion to the intensity with which others try to prove him wrong"
James Mason from the movie "Heaven Can Wait".
Steve Jaros
what's
> being debated.
What would you consider to be evidence of a clutch hitter?
TJNawrocki
: In a total of 77 postseason games, Mr. October hit .278 with 18 HR and 48 HR
in
: a 77 games; roughly half a season of LCS and WS games. If you double those
: totals to approximate a 162 game season, you have .278 with 36 HR and 96 RBI
in
: 154 games; his career average per 162 games is .262 BA with 32 HR and 98 RBI.
: In other words his PS performance compares favorably to his regular season
: stats although the BA is a little bit higher.
Derek Jeter is the same way. Coming into this season, his postseason averages
couldn't really be much closer to his career averages: .314/.381/.469 in the
postseason, and .317/.389/.463 in the regular season. It's a good record, of
course, especially considering he's generally facing tougher pitching in the
postseason, but it hardly justifies Tim McCarver's drooling all over him.
Tom Nawrocki
igor eduardo küpfer
<TrojanFamily...@trojannutcase.com> wrote in
<K3Rlb.90608$sp2.11244@lakeread04>:
>> Nobody says clutch hits don't exist. The existence of clutch hitters is
>what's
>> being debated.
>
>What would you consider to be evidence of a clutch hitter?
Players who postseason performances are better than can be expected by chance.
Ron Johnson
Steve Jaros <TrojanFamily...@trojannutcase.com> wrote:
>> Nobody says clutch hits don't exist. The existence of clutch hitters is
>what's
>> being debated.
>
>What would you consider to be evidence of a clutch hitter?
Statistical significance.
I'm prepared to argue that Paul Molitor and Tony Fernandez meet
a definition of clutch hitter (though I don't insist that they
are -- I merely note that they meet a particular test)
Between 1984 and 1998 Paul Molitor hit .312 overall and .345 with
RISP. Sample size is just under 2,000 AB with RISP. Standard
deviation in this kind of sample size is around 9-10 points.
Between 1984 and 1999 Tony Fernandez hit .287 overall and .319 with
RISP. Just over 1,850 AB. Nobody else has reached statistical
significance. Though Pat Tabler and Scott Fletcher were close.
Nobody reached statistical significance as a choker. The worst I
can find here are Walt Weiss, Jim Gantner and Phil Bradley.
--
RNJ
SavoyBG
>Nobody else has reached statistical
>significance.
What about Eddie Murray?
Bruce Grossberg
Ben
> can find here are Walt Weiss, Jim Gantner and Phil Bradley.
that surprises me. it seems that no one reaches mlb unless he is doing his
dead-level best every pa, so the idea that some guys are doing less than their
best except in "clutch" situations doesn't make sense. but you hear mlb players
blame their slumps on "pressing" all the time, and i think i've seen normally
patient hitters swinging at bad pitches when the atmosphere was charged. so,
intuitively, i am much more willing to consider choking than clutch.
then again, i'm a braves fan.
--
Ben B
Douglas T. (Doug) Massey
I heard many folks here say that there's no such thing as choke hitters,
since players with that tendency have been weeded out by pressure situations
in high school, college, and the minors long before reaching the majors.
But I think it's possible that a hitter never really meets a pressure
situation before getting to the majors, simply because he dominates every
league leading up to the majors. Perhaps it's one type of pressure to
be batting with the bases loaded and two outs in the ninth inning of a
high school game where you know that you're far better than any guy
who might happen to be on the mound, but it's another type of pressure
to have to do so against MLB pitching -- where you know the guy is
basically your equal. So the fear of failure really only pops up once
he reaches the major league . . .
Just thinking . . .
Doug
--
-----------------------------------------------------------------------
___, Doug Massey, ASIC Digital Logic Designer
\o IBM Microelectronics Division, Burlington, Vermont |>
| Phone: (802)769-7095 t/l: 446-7095 fax: x6752 |
/ \ |
. My homepage: http://doug.obscurestuff.com (|)
Alan S. Wales
>Nobody reached statistical significance as a choker. The worst I
>can find here are Walt Weiss, Jim Gantner and Phil Bradley.
Torii Hunter 2003
runners on: .228/.289/.429
RISP .246/.306/.445
no one on: .271/.334/.473
--
"When you argue with a fool be sure he is not similarly occupied."
See how the Pros get more POWER!
http://www.powrwrap.com/press.htm
Ron Johnson
SavoyBG <sav...@aol.com> wrote:
>>From: joh...@ccrs.nrcan.gc.ca (Ron Johnson)
>
>>Nobody else has reached statistical
>>significance.
>
>What about Eddie Murray?
He meets a different definition. The only hitter I've found
in a long career who has better numbers in late/close than
overall.
--
RNJ
Steve Jaros
news:0qsfpv89n94f8q651...@4ax.com...
> On Thu, 23 Oct 2003 08:52:53 -0500, "Steve Jaros"
> <TrojanFamily...@trojannutcase.com> wrote in
> <K3Rlb.90608$sp2.11244@lakeread04>:
>
> >> Nobody says clutch hits don't exist. The existence of clutch hitters is
> >what's
> >> being debated.
> >
> >What would you consider to be evidence of a clutch hitter?
>
> Players who postseason performances are better than can be expected by
chance.
If 'by chance' you mean by comparison to their regular season performance, i
don't see how that would work. For example, a hit that wins game 155 when
your team is in a very tight playoff race is surely a clutch hit, whereas a
hit that comes in game 5 of the WS when your team is winning the game 13-1
in the 8th inning isn't...
Ron Johnson
Alan S. Wales <powr...@aol.compost> wrote:
>>joh...@ccrs.nrcan.gc.ca (Ron Johnson)
>
>>Nobody reached statistical significance as a choker. The worst I
>>can find here are Walt Weiss, Jim Gantner and Phil Bradley.
>
>Torii Hunter 2003
>
>runners on: .228/.289/.429
>RISP .246/.306/.445
>no one on: .271/.334/.473
That's nothing close to being significant.
Standard deviation of batting stats over a month's regular play:
49 points of BA, 57 point of OBP and 117 points of SLG
That's the results of comparing players in consecutive months
that they had at leat 90 PAs. Done this way to try and cut
down on the impact that injuries would have on the study.
Single season RISP numbers basically can't tell you anything
of interest. (And have next to nothing to do with run scoring
in any case)
--
RNJ
igor eduardo küpfer
<TrojanFamily...@trojannutcase.com> wrote in
<siWlb.90643$sp2.53614@lakeread04>:
That's why I said that clutch hits exist, but clutch hitting ability is
debatable. If a player is a clutch hitter, then he has the ability to perform
better than normal under clutch circumstances. As Ron Johnson points out in
another post, only two players have hit significantly better with RISP. This
isn't the only definition of a clutch situation, but isn't it odd that only
two players are "clutch" with runners in scoring position? Almost like the
phenomenon doesn't exist...
Going to basketball, 211 players have attempted more 200 or more free throws
in the playoffs. No player has hit free throws significantly better in the
playoffs than he did in the regular season. Not one. (On the other end, one
player hit significantly worse in the playoffs: Wilt Chamberlain. And we all
know what a choker he was.)
Craig Richardson
wrote:
>Derek Jeter is the same way. Coming into this season, his postseason averages
>couldn't really be much closer to his career averages: .314/.381/.469 in the
>postseason, and .317/.389/.463 in the regular season. It's a good record, of
>course, especially considering he's generally facing tougher pitching in the
>postseason, but it hardly justifies Tim McCarver's drooling all over him.
My favorite was the graphic which showed that Jeter has gotten on base
in about 90% of his postseason starts. A guy with a .400 OBA will
miss out in 4 or 5 plate appearances something like ... 10% of the
time.
While he deserves props for (1) being consistent across multiple
series, even multiple seasons, and (2) keeping a .400 OBA against
better-than-regular-season pitching, it's not like it's an
earth-shattering departure indicating incipent Godhood.
--Craig
--
I start to wish Bob Melvin would walk out to the mound, ask Freddy if he
was injured, and then kick him in the balls so he can call in an
emergency replacement from the bullpen --Derek Zumsteg in BP, 5/13/2003
Doug Norris
>It's hard to believe that there are still some people who don't 'believe' in
>the notion of clutch hitting.
Right. That's funny, isn't it. There's also still some people that don't
believe Elvis is dead, either. Can you believe it?!
Doug
nob...@junk.min.net
>Derek Jeter is the same way. Coming into this season, his postseason
>averages couldn't really be much closer to his career averages:
>.314/.381/.469 in the postseason, and .317/.389/.463 in the regular
>season. It's a good record, of course, especially considering he's
>generally facing tougher pitching in the postseason, but it hardly
>justifies Tim McCarver's drooling all over him.
They also think he's a good shortstop, which he isn't.
Alan
--
---------------------------------------------------------------------
**** Please use address alanh(at)min.net to reply via e-mail. ****
Posted using registered MR/2 ICE Newsreader #564 and eComStation 1.1
---------------------------------------------------------------------
Doug Norris
>"Steve Jaros" <TrojanFamily...@trojannutcase.com> writes:
Of course, I mean "...believe Elvis isn't dead." I'm such a dope.
Doug
David J. Grabiner
> In article <K3Rlb.90608$sp2.11244@lakeread04>,
> Steve Jaros <TrojanFamily...@trojannutcase.com> wrote:
> >> Nobody says clutch hits don't exist. The existence of clutch hitters is
> >what's
> >> being debated.
> >
> >What would you consider to be evidence of a clutch hitter?
>
> Statistical significance.
>
> I'm prepared to argue that Paul Molitor and Tony Fernandez meet
> a definition of clutch hitter (though I don't insist that they
> are -- I merely note that they meet a particular test)
>
> Between 1984 and 1998 Paul Molitor hit .312 overall and .345 with
> RISP. Sample size is just under 2,000 AB with RISP. Standard
> deviation in this kind of sample size is around 9-10 points.
For a .312 hitter, the variance of his batting average in 2000 AB with
RISP is (.312)*(.688)/2000=.000107. The variance of his batting average
in 6000 AB without RISP is (.312)*(.688)/6000=.0000036. Adding these
variances and taking the square root gives a standard deviation of 12
points, not 10 points.
The probability that a random player will exceed the expected value by
at least 33 points, when the standard deviation is 12 points, is .003.
In other words, if you looked at all 1998 regulars with long careers,
and there were no ability, this is about where you would expect the top
player to be, given that there are about 250 players. It is not
statistically significant.
Note that this does not disprove the existence of clutch hitting. If
there is an ability to hit better with runners in scoring position, and
the standard deviation of that ability is 5 points (one hit every other
year), then the actual standard deviation over a career of Molitor's
length is 13 points. But such an ability wouldn't mean much; the top 2%
of clutch hitters would have an ability to get one extra hit a year.
--
David Grabiner, grab...@alumni.princeton.edu, http://remarque.org/~grabiner
Baseball labor negotiations FAQ: http://remarque.org/~grabiner/laborfaq.html
Shop at the Mobius Strip Mall: Always on the same side of the street!
Klein Glassworks, Torus Coffee and Donuts, Projective Airlines, etc.
Roger Moore
>While he deserves props for (1) being consistent across multiple
>series, even multiple seasons, and (2) keeping a .400 OBA against
>better-than-regular-season pitching, it's not like it's an
>earth-shattering departure indicating incipent Godhood.
Except that he hasn't been that consistent. (Nor, for that matter has he
kept up a .400 OBA: it's .381 in post-season play and .360 in Series play
going into 2003.) He's had some very good series and some very bad series.
Copying liberally from Baseball-Reference.com, here are his
series-by-series records
Year Series BA OBA SLG
1996 ALDS .412 .412 .471
1996 ALCS .417 .417 .625
1996 WS .250 .375 .250
1997 ALDS .333 .417 .667
1998 ALDS .111 .273 .111
1998 ALCS .200 .259 .320
1998 WS .353 .450 .353
1999 ALDS .455 .538 .727
1999 ALCS .350 .409 .550
1999 WS .353 .389 .412
2000 ALDS .211 .286 .211
2000 ALCS .318 .464 .591
2000 WS .409 .480 .864
2001 ALDS .444 .474 .500
2001 ALCS .118 .211 .118
2001 WS .148 .148 .259
2002 ALDS .500 .556 .875
2003 ALDS .429 .556 .643
2003 ALCS .233 .281 .400
2003 WS .333 .368 .500
It looks to me as though sometimes he hits like a hero (2000 World
Series, 1999 Division Series) and sometimes disappears from the face of
the earth (1998 Division Series, 2001 Championship Series). Sometimes
he's come up huge in Series his team has won (1999 Division Series) and
sometime in Series his team has lost (2002 Division Series). Sometimes
he's sucked in Series his team has won (2000 Division Series) and
sometimes in Series his team has lost (2001 World Series).
--
Roger Moore | Master of Meaningless Trivia | (r...@alumni.caltech.edu)
I believe there are more instances of the abridgement of freedom of the
people by gradual and silent encroachments by those in power than by
violent and sudden usurpations. -- James Madison
Bob Roman
wrote:
>that surprises me. it seems that no one reaches mlb unless he is doing his
>dead-level best every pa, so the idea that some guys are doing less than their
>best except in "clutch" situations doesn't make sense.
This is the argument favored by RSB'ers, but I still say it is
physiologically inaccurate. Clutch hitting, if it exists, is not a
matter of personal volition -- trying harder, "cranking it up," etc.
Certain situations simply arouse greater nervous energy.
It's like with runners. Even Olympic-level runners will run faster in
competition than they will against the clock. The effect appears to
be outside of conscious awareness, and is usually called social
facilitation.
You are correct to suggest that clutch and choke are two sides of the
same coin. The energy manifested through social facilitation can help
or hinder performance, basically as a function of the talents of the
performer. Dominant, well-practiced behaviors will be helped; weak,
underlearned behaviors will be hindered.*
I'm not saying clutch hitters do exist at the MLB level -- the
evidence says they generally do not. But to me the more interesting
issue is why clutch hitters are not found more often. In theory,
clutch hitting should be most common among the talent levels of
MLB-caliber players.
Maybe there's such a thing as clutch pitching.
* Zajonc, RB, & Sales, SM (1966). Social facilitation of dominant and
subordinate responses. Journal of Experimental Social Psychology, 2,
160-168
Bob Roman
Steve Cutchen
<joh...@ccrs.nrcan.gc.ca> wrote:
Even here, the explanation could easily be a different approach taken
with RISP versus coming up with 2 outs and no one on. For example, in
the latter case, a batter might be much more likely to swing less
carefully in order to perhaps clear the fence.
There's no way to break out the batter's different philosophy in
different scenarios and how that affects whether he's a "clutch
hitter."
Besides, wouldn't "clutch hitter" be synonymous with "slacker"? Why
wouldn't a batter try hard all of the time?
Craig Richardson
(Roger Moore) wrote:
>Craig Richardson <crichar...@worldnet.att.net> writes:
>
>>While he deserves props for (1) being consistent across multiple
>>series, even multiple seasons, and (2) keeping a .400 OBA against
>>better-than-regular-season pitching, it's not like it's an
>>earth-shattering departure indicating incipent Godhood.
>
>Except that he hasn't been that consistent. (Nor, for that matter has he
>kept up a .400 OBA: it's .381 in post-season play and .360 in Series play
>going into 2003.) He's had some very good series and some very bad series.
Even so, he's still been on base in 86 of 95 games, or something like
that, so even on off nights he still does /something/.
It's remarkable, i.e. worthy of remark. I'm not sure it's
outstanding, and it certainly isn't an indication of excellence (above
and beyond his normal ability, which might or might not be excellent).
Craig Richardson
Norris) wrote:
ObMenInBlack: "You do know Elvis is dead, don't you?" "Nope. He just
went home."
ObSneakers: "Are you suggesting the NSA killed Kennedy?" "No - they
shot him, but they didn't kill him. He's still alive!"
Steve Cutchen
<tjnaw...@aol.compost> wrote:
Part of what gives the impression of a "clutch hitter" is that better
hitters have a higher chance to be successful in each clutch situation
(and each non-clutch situation).
They succeed more in clutch situations simply because they succeed more
in all situations.
Steve Cutchen
<Robert...@hotmail.com> wrote:
Perhaps to get to the MLB level, a player has to have mastered the
effects of social facilitation to his benefit pretty much all of the
time. IOWs, he's always taking pretty darned good advantage of his
nervous energy, and it is a pretty even effect, even when comparing
regular season games to playoffs.
> Maybe there's such a thing as clutch pitching.
Ask BH Kim... :-)
Steve Jaros
performance, i
> >don't see how that would work. For example, a hit that wins game 155 when
> >your team is in a very tight playoff race is surely a clutch hit, whereas
a
> >hit that comes in game 5 of the WS when your team is winning the game
13-1
> >in the 8th inning isn't...
> >
>
> That's why I said that clutch hits exist, but clutch hitting ability is
> debatable. If a player is a clutch hitter, then he has the ability to
perform
> better than normal under clutch circumstances.
So what we need to do is define clutch circumstances...
Steve Jaros
> there is an ability to hit better with runners in scoring position, and
> the standard deviation of that ability is 5 points (one hit every other
> year), then the actual standard deviation over a career of Molitor's
> length is 13 points. But such an ability wouldn't mean much; the top 2%
> of clutch hitters would have an ability to get one extra hit a year.
Right, which is why "statistical significance" isn't as useful as
'practical' significance.
I don't see much value in comparing a guy's regular average and his
post-season or RISP average. What's relevant is comparing him to the league
average. Clutch hitting should mean "the guy gets significantly more hits
than most other guys do when the situation really matters".
For example, if a player hit .390 during the regular season and .380 in
crucial post-season or pennant-race regular season situations, *and* the
league average in these situations is say .290, i'd call him a clutch
hitter, because .380 is by comparison to the rest of the league, a great
clutch BA, even though it's lower than his personal non-clutch situation
BA...
brink
"Bob Roman" <Robert...@hotmail.com> wrote in message
news:3f98a162...@news.supernews.com...
> On Thu, 23 Oct 2003 12:50:56 -0400, "Ben" <bbla...@hotmail.com>
> wrote:
>
> >that surprises me. it seems that no one reaches mlb unless he is doing
his
> >dead-level best every pa, so the idea that some guys are doing less than
their
> >best except in "clutch" situations doesn't make sense.
>
> This is the argument favored by RSB'ers, but I still say it is
> physiologically inaccurate. Clutch hitting, if it exists, is not a
> matter of personal volition -- trying harder, "cranking it up," etc.
> Certain situations simply arouse greater nervous energy.
>
> It's like with runners. Even Olympic-level runners will run faster in
> competition than they will against the clock. The effect appears to
> be outside of conscious awareness, and is usually called social
> facilitation.
this is an interesting theory, but remember that running against a clock or
even against other runners is a non-zero sum game while hitter vs. a pitcher
*is*. let's say competition does make runners run faster than simply
"against the clock"--in that case, all eight runners in an olympic race may
very well perform better than under ordingary circumstances. but in
baseball, for every winner there's a loser. when a batter gets a hit, he
wins while the pitcher "loses." and vice versa on a flyout or strikeout.
so your theory runs into itself in that pitchers, too should be performing
better in the postseason since they're probably among the best in the league
already and they'll be amped up by the increased level of play and
importance of the games.
so wouldn't any physiological gains by the batter tend to be mitigated by
the physiological gains realized by the pitcher?
>
> You are correct to suggest that clutch and choke are two sides of the
> same coin. The energy manifested through social facilitation can help
> or hinder performance, basically as a function of the talents of the
> performer. Dominant, well-practiced behaviors will be helped; weak,
> underlearned behaviors will be hindered.*
>
> I'm not saying clutch hitters do exist at the MLB level -- the
> evidence says they generally do not. But to me the more interesting
> issue is why clutch hitters are not found more often. In theory,
> clutch hitting should be most common among the talent levels of
> MLB-caliber players.
well there's been a lot of talk about how jeter puts up essentially
identical numbers in the postseason to his regular season production. and
actually what's been overlooked is that this in itself is perhaps
remarkable--after all, he's putting up what were already very good numbers
against "weaker" competition in the regular season on average (when he gets
to face the devil rays, tigers, rangers, and their ilk on a regular basis)
and somehow he's been able to *maintain* that type of production when the
pitchers are unquestionably tougher to hit once the playoffs arrive.
> Maybe there's such a thing as clutch pitching.
and maybe it cancels out clutch hitting.
> * Zajonc, RB, & Sales, SM (1966). Social facilitation of dominant and
> subordinate responses. Journal of Experimental Social Psychology, 2,
> 160-168
a post with a footnote... i like that. ;-)
>
> Bob Roman
Craig Richardson
wrote:
>In article <3f98a162...@news.supernews.com>, Bob Roman
><Robert...@hotmail.com> wrote:
>
>> I'm not saying clutch hitters do exist at the MLB level -- the
>> evidence says they generally do not. But to me the more interesting
>> issue is why clutch hitters are not found more often. In theory,
>> clutch hitting should be most common among the talent levels of
>> MLB-caliber players.
>
>Perhaps to get to the MLB level, a player has to have mastered the
>effects of social facilitation to his benefit pretty much all of the
>time. IOWs, he's always taking pretty darned good advantage of his
>nervous energy, and it is a pretty even effect, even when comparing
>regular season games to playoffs.
My supposition is that every MLB at-bat is a clutch at-bat. Batting
at a high level is unlike most every other experience in team sports -
especially in that there's nothing one can actively do to improve
one's performance. Hustle, adrenaline, etc. - none of that helps, and
some of it actively hurts. Like golf, concentration and how it allows
one access to muscle memory through training, comes to the fore.
>> Maybe there's such a thing as clutch pitching.
>
>Ask BH Kim... :-)
OTOH, some pitchers - not the Jamie Moyers but the Armando Benitezes,
i.e. the ones who get by with tools rather than command, might be
helped by adrenaline. Yeah, yeah, it's a continuum - but I still
believe that command pitchers are helped more by pressure "chokers" as
opposed to "power" pitchers, who have more chance to help themselves.
igor eduardo küpfer
<TrojanFamily...@trojannutcase.com> wrote in
<ed0mb.91093$sp2.41751@lakeread04>:
I'm game. IIRC there have been a number of "clutch" definitions in MBL
tested. Basketball is the game I know best -- playoff vs reg season
performance differences have so far been indistinguishable from chance. Play
by play data will make this an topic that can be studied in more depth in the
years to come, so we may yet see some evidence that clutch ability exists.
igor eduardo küpfer
<TrojanFamily...@trojannutcase.com> wrote in
<Yj0mb.91094$sp2.13787@lakeread04>:
<snip>
>Right, which is why "statistical significance" isn't as useful as
>'practical' significance.
They are both useful and important. A career .300 hitter goes 4-5 in one game.
The practical significance is an increase of .500 over his career average. The
statistical significance is nil. Which is more useful: he tripled his ability?
Or he had a flukey game? Maybe some combination?
<...>
Bob Roman
>
>"Bob Roman" <Robert...@hotmail.com> wrote...
>> Maybe there's such a thing as clutch pitching.
>
>and maybe it cancels out clutch hitting.
Yes, that was my point as well.
The only area of baseball that is not a zero-sum game is fielding. I
don't know the research, but perhaps we could find evidence for
"clutch fielding."
>> * Zajonc, RB, & Sales, SM (1966). Social facilitation of dominant and
>> subordinate responses. Journal of Experimental Social Psychology, 2,
>> 160-168
>
>a post with a footnote... i like that. ;-)
It's not so common now, but back in the mid-90s it was practically
required. You couldn't make a statement of fact here without someone
saying "citation?" or "reference, please."
Bob Roman
Steve Jaros
> On Thu, 23 Oct 2003 21:42:01 -0500, "Steve Jaros"
> <TrojanFamily...@trojannutcase.com> wrote in
> <Yj0mb.91094$sp2.13787@lakeread04>:
>
> <snip>
>
> >Right, which is why "statistical significance" isn't as useful as
> >'practical' significance.
>
> They are both useful and important. A career .300 hitter goes 4-5 in one
game.
> The practical significance is an increase of .500 over his career average.
The
> statistical significance is nil. Which is more useful: he tripled his
ability?
> Or he had a flukey game? Maybe some combination?
Good point. Surely, to be called a 'clutch hitter', the guy has to do it
enough times (i.e., we have to have a large enough 'sample') for us to rule
out 'fluke'....
Douglas T. (Doug) Massey
"Steve Jaros" <TrojanFamily...@trojannutcase.com> writes:
>
> I don't see much value in comparing a guy's regular average and his
> post-season or RISP average. What's relevant is comparing him to the league
> average. Clutch hitting should mean "the guy gets significantly more hits
> than most other guys do when the situation really matters".
But that's not clutch hitting -- that's "good hitting".
> For example, if a player hit .390 during the regular season and .380 in
> crucial post-season or pennant-race regular season situations, *and* the
> league average in these situations is say .290, i'd call him a clutch
> hitter, because .380 is by comparison to the rest of the league, a great
> clutch BA, even though it's lower than his personal non-clutch situation
> BA...
Again, you're not identifying anything interesting, other than "who is
a good hitter?".
By your definition, the best clutch hitters would be the best hitters.
Well, duh. I'm not saying your definition isn't valid -- but it is
trivial.
Most people consider "clutch hitting" to be the ability to hit better
in certain "clutch" scenarios than in non-clutch scenarios -- for some
definition of clutch scenarios. That's how a poor hitter could conceivably
still be clutch and a great hitter could still conceivably be a choker.
Doug
--
-----------------------------------------------------------------------
___, Doug Massey, ASIC Digital Logic Designer
\o IBM Microelectronics Division, Burlington, Vermont |>
| Phone: (802)769-7095 t/l: 446-7095 fax: x6752 |
/ \ |
. My homepage: http://doug.obscurestuff.com (|)
Perry Sailor
"Steve Jaros" <TrojanFamily...@trojannutcase.com> wrote in message
news:Yj0mb.91094$sp2.13787@lakeread04...
> I don't see much value in comparing a guy's regular average and his
> post-season or RISP average. What's relevant is comparing him to the
league
> average. Clutch hitting should mean "the guy gets significantly more hits
> than most other guys do when the situation really matters".
>
> For example, if a player hit .390 during the regular season and .380 in
> crucial post-season or pennant-race regular season situations, *and* the
> league average in these situations is say .290, i'd call him a clutch
> hitter, because .380 is by comparison to the rest of the league, a great
> clutch BA, even though it's lower than his personal non-clutch situation
> BA...
Well, you're entitled to define it any way you want, and it's good that you
clarified your definition, but 99 people out of 100, I suspect, define it in
comparison to the player's own established overall performance, as a matter
of "rising to the occasion." Your hypothetical player (call him George
Brett, say), isn't clutch, he's just great. In other words, defining it
your way, "clutch" adds nothing to what we already know about the player,
and thus has no meaning at all.
Perry
Steve Jaros
news:bnbdgt$92m$1...@peabody.colorado.edu...
>
> "Steve Jaros" <TrojanFamily...@trojannutcase.com> wrote in
message
> news:Yj0mb.91094$sp2.13787@lakeread04...
> > I don't see much value in comparing a guy's regular average and his
> > post-season or RISP average. What's relevant is comparing him to the
> league
> > average. Clutch hitting should mean "the guy gets significantly more
hits
> > than most other guys do when the situation really matters".
> >
> > For example, if a player hit .390 during the regular season and .380 in
> > crucial post-season or pennant-race regular season situations, *and* the
> > league average in these situations is say .290, i'd call him a clutch
> > hitter, because .380 is by comparison to the rest of the league, a great
> > clutch BA, even though it's lower than his personal non-clutch situation
> > BA...
>
> Well, you're entitled to define it any way you want, and it's good that
you
> clarified your definition, but 99 people out of 100, I suspect, define it
in
> comparison to the player's own established overall performance, as a
matter
> of "rising to the occasion."
Perhaps so, but I don't see why they would define it that way. A guy hits
.200 during non-clutch situations and .240 in clutch situations is a 'clutch
hitter'? He's a bad hitter in both situations, just less-bad in the
latter...
> Your hypothetical player (call him George
> Brett, say), isn't clutch, he's just great. In other words, defining it
> your way, "clutch" adds nothing to what we already know about the player,
> and thus has no meaning at all.
Actually, it does - it tells us he's great hitting in the clutch, which is a
different situation from non-clutch hitting and one we don't know about
until we calculate his 'clutch' BA.
He's just great at both.
Dale Hicks
Robert...@hotmail.com says...
> On Thu, 23 Oct 2003 23:55:53 -0700, "brink" <for...@about.it> wrote:
>
> >"Bob Roman" <Robert...@hotmail.com> wrote...
> >> Maybe there's such a thing as clutch pitching.
> >and maybe it cancels out clutch hitting.
>
> Yes, that was my point as well.
One thing to keep in mind that in some definitions of clutch, the
pitching's altered simply due to the pitcher being forced to pitch from
the stretch, where you can't get enough leverage on the ball. I have no
clue if that affects either of your arguments. Perhaps certain batters
can be baffled by the trick windup pitchers?
> The only area of baseball that is not a zero-sum game is fielding. I
> don't know the research, but perhaps we could find evidence for
> "clutch fielding."
If you go with Voros' Conjecture then pitchers lose a certain amount of
control when it leaves their hand, more than batters lose.
> >a post with a footnote... i like that. ;-)
>
> It's not so common now, but back in the mid-90s it was practically
> required. You couldn't make a statement of fact here without someone
> saying "citation?" or "reference, please."
Citation?
--
Cranial Crusader dgh 1138 at bell south point net
Alan S. Wales
>>powr...@aol.compost> wrote:
>>> joh...@ccrs.nrcan.gc.ca (Ron Johnson)
>>>Nobody reached statistical significance as a choker. The worst I
>>>can find here are Walt Weiss, Jim Gantner and Phil Bradley.
>>Torii Hunter 2003
>>
>>runners on: .228/.289/.429
>>RISP .246/.306/.445
>>no one on: .271/.334/.473
>That's nothing close to being significant.
Tell that to the Twins and their fans!
>Standard deviation of batting stats over a month's regular play:
>
>49 points of BA, 57 point of OBP and 117 points of SLG
>
>That's the results of comparing players in consecutive months
>that they had at leat 90 PAs. Done this way to try and cut
>down on the impact that injuries would have on the study.
>
>Single season RISP numbers basically can't tell you anything
>of interest. (And have next to nothing to do with run scoring
>in any case)
--
"When you argue with a fool be sure he is not similarly occupied."
See how the Pros get more POWER!
http://www.powrwrap.com/press.htm
Perry Sailor
"Ron Johnson" <joh...@ccrs.nrcan.gc.ca> wrote in message
news:bn8uo4$a...@gcpdb.ccrs.nrcan.gc.ca...
> In article <K3Rlb.90608$sp2.11244@lakeread04>,
> Steve Jaros <TrojanFamily...@trojannutcase.com> wrote:
> >> Nobody says clutch hits don't exist. The existence of clutch hitters is
> >what's
> >> being debated.
> >
> >What would you consider to be evidence of a clutch hitter?
>
> Statistical significance.
>
> I'm prepared to argue that Paul Molitor and Tony Fernandez meet
> a definition of clutch hitter (though I don't insist that they
> are -- I merely note that they meet a particular test)
And the thing is, with statistical significance set at the conventional .05,
2.5% of players will look like real clutch hitters (and 2.5% as chokers)
EVEN IF THERE'S NO SUCH THING. That's what .05 means in significance
tests -- that 5% of the time, you'll get statistical significance, even if
there's no real effect. The logic is, "okay, if there were no effect, 5% of
the time we'd get this result, 95% of the time we wouldn't. We got this
result, therefore there's an effect." Hardly a watertight case, so
statistical significance alone is not infallible. Far better is statistical
significance plus replication in independent samples. You get that, you're
getting somewhere.
Perry
Ron Johnson
Doesn't *seem* to be a big factor.
I did a study of every player with 1,000+ PAs with RISP between 1984 and
1998.
They hit .275/.346/.432/.340 overall and .278/.369/.430/.345 with RISP.
(The last number being OBP with IBB removed. Didn't think to remove
SF and I don't feel like re-doing this. Won't change the essential
conclusion.)
There is however only an 86% correlation between BA with RISP and
overall BA. Quite strong, but it's evidence that a little something
is probably going on with RISP.
And that it would be right tough to detect in the results of any
given player. And that it probably doesn't matter much.
One reason so many people overrate the importance of hitting with
RISP is they assume that a single with RISP is an RBI. In fact
the runner on second only scores on just over 60% of all singles to
either left or right and just over 80% of the time on a single to
center. (and of course about never on an infield single)
Yeah, I was surprised that the figure was that low.
>
>Besides, wouldn't "clutch hitter" be synonymous with "slacker"? Why
>wouldn't a batter try hard all of the time?
That's my argument too.
--
RNJ
David J. Grabiner
> I'm not saying clutch hitters do exist at the MLB level -- the
> evidence says they generally do not. But to me the more interesting
> issue is why clutch hitters are not found more often. In theory,
> clutch hitting should be most common among the talent levels of
> MLB-caliber players.
> Maybe there's such a thing as clutch pitching.
It may be that there is an ability, but all MLB hitters have the ability
at about the same level. That would make it impossible to identify the
top clutch hitters.
David J. Grabiner
But it's something that would be meaningful in evaluating the player.
If he hits .200 in non-clutch situations and .240 in clutch situations,
and clutch situations are 1/6 of the time, then he is an overall .207
hitter but he is as valuable as a .215 hitter because clutch situations
are twice as important to winning games as non-clutch situations. He
may still not be a good hitter, but that extra value might be enough to
give him a regular job as catcher, or to avoid pinch-hitting for him
when he comes up in the eighth inning of a tie game.
Conversely, if there is a player who hits .350 in non-clutch situations
and .300 in clutch situations for an overall .342, he is still a good
hitter, but if that is an ability, then MVP voters should consider it.
(I don't think there is any player with an ability that strong in either
direction. All of the studies of clutch hitting have found either no
effect or a very small effect; if players had abilities this extreme, we
could identify a larger effect.)
David J. Grabiner
> I did a study of every player with 1,000+ PAs with RISP between 1984 and
> 1998.
>
> They hit .275/.346/.432/.340 overall and .278/.369/.430/.345 with RISP.
> (The last number being OBP with IBB removed. Didn't think to remove
> SF and I don't feel like re-doing this. Won't change the essential
> conclusion.)
>
> There is however only an 86% correlation between BA with RISP and
> overall BA. Quite strong, but it's evidence that a little something
> is probably going on with RISP.
What would the correlation be if there were no ability, and you just
picked a random sample of 1/4 of each hitter's AB as his RISP ones?
I did a study back in 1991, and estimates that the top clutch hitters
hit .006 better than expected with RISP, but that was only one standard
deviation. (And it was of no baseball significance; .006 is less than
one hit a year.)
Kenny1111
grab...@alumni.princeton.edu says...
> (I don't think there is any player with an ability that strong in either
> direction. All of the studies of clutch hitting have found either no
> effect or a very small effect; if players had abilities this extreme, we
> could identify a larger effect.)
Speaking of the "either direction" part, did you in your study, or could
you refer me to a study, in which someone showed both ways -- i.e., that
"clutch" hitting and "choking" both don't exist (to a statistically
significant level)?
Because I recently got in an argument with someone about this, and he
was able to accept my claims that clutch hitting (hitting better with
RISP) doesn't exist, but then put forth the claim that clutch hitters
are those that do the same, while most players do worse. Now, I'm
fairly sure this is false also, but having proof would be good :-)
My reply was that studies with the goal of disproving "clutch" hitting
(as opposed to choking) have shown the the distribution of hitting with
RISP compared to normal situations is no different than what one would
expect from total randomness, and that this works both ways. But I also
seem to remember that most of what I read took more of a better-than-
normal focus, and don't have a study to support my contention.
Steve Jaros
that
> > you
> > > clarified your definition, but 99 people out of 100, I suspect, define
it
> > in
> > > comparison to the player's own established overall performance, as a
> > matter
> > > of "rising to the occasion."
> >
> > Perhaps so, but I don't see why they would define it that way. A guy
hits
> > .200 during non-clutch situations and .240 in clutch situations is a
'clutch
> > hitter'? He's a bad hitter in both situations, just less-bad in the
> > latter...
>
> But it's something that would be meaningful in evaluating the player.
> If he hits .200 in non-clutch situations and .240 in clutch situations,
> and clutch situations are 1/6 of the time, then he is an overall .207
> hitter but he is as valuable as a .215 hitter because clutch situations
> are twice as important to winning games as non-clutch situations. He
> may still not be a good hitter, but that extra value might be enough to
> give him a regular job as catcher, or to avoid pinch-hitting for him
> when he comes up in the eighth inning of a tie game.
Yes. I didn't mean to claim there wasn't potential baseball utlity in that
difference. As you note it could mean that he might be the best option a
team has in a given 'clutch' situation whereas he wouldn't be the best
option in a non-clutch situation. But that still doesn't mean he's a 'clutch
hitter', if by that we mean a guy who is good in the clutch - defined to
mean how the player's clutuch performance compares to a standard ('average'
clutch hitting BA), not his own personal non-clutch performance.
David J. Grabiner
> In article <uoew6f...@alumni.princeton.edu>,
> grab...@alumni.princeton.edu says...
>
> > (I don't think there is any player with an ability that strong in either
> > direction. All of the studies of clutch hitting have found either no
> > effect or a very small effect; if players had abilities this extreme, we
> > could identify a larger effect.)
>
> Speaking of the "either direction" part, did you in your study, or could
> you refer me to a study, in which someone showed both ways -- i.e., that
> "clutch" hitting and "choking" both don't exist (to a statistically
> significant level)?
>
> Because I recently got in an argument with someone about this, and he
> was able to accept my claims that clutch hitting (hitting better with
> RISP) doesn't exist, but then put forth the claim that clutch hitters
> are those that do the same, while most players do worse. Now, I'm
> fairly sure this is false also, but having proof would be good :-)
Batting average with RISP is the same as without RISP. In the late
innings of close games, hitting goes down, but this may be due to the
use of relief aces.
If there are some hitters with clutch ability significantly better or
worse than an average hitter, then we could identify the ability; the
best clutch hitters in the past would continue to hit well this year,
and the worst chokers would continue to choke. If there is no ability,
then there will be no correlation between past and current clutch
performance. Even if all the deviations were on one side (chokers but
no clutch hitters), we would just find an effect that was half as
strong.
Here's a study I did ten years ago in which I looked for this effect,
and found essentially no correlation, along with some possible reasons
that there should be a very slight correlation even if there is no
clutch ability. (For example, with the use of spot left-handed
relievers, good left-handed hitters with large platoon splits are less
likely to have the platoon advantage in the late innings of close
games.)
http://remarque.org/~grabiner/fullclutch.txt
Kenny1111
grab...@alumni.princeton.edu says...
Thanks very much. Just one question, though. When Voros did his DIPS
research, he found little correlation on BA on balls in play by doing a
similar year-to-year correlation study. Then Woolner(?) and Tippett did
studies where they looked at larger periods of time, and seemed to find
that a pitcher probably does control what happens on balls in play to a
great extent than Voros may have thought, although still far less than
our intuition may have led us to believe before Voros released his
results.... Now, not to criticize your study in any way, but is there
any possibility that the same thing can be true with your study? (I am
by know means trying to say that I believe clutch hitting exists,
because I don't, and have seen various studies over the years showing so
... I just want the best "ammunition" as possible to prove my case for
the choker thing, even though I'm probably being neurotic and overally
picky :-) )
Ron Johnson
David J. Grabiner <grab...@alumni.princeton.edu> wrote:
>joh...@ccrs.nrcan.gc.ca (Ron Johnson) writes:
>
>> In article <K3Rlb.90608$sp2.11244@lakeread04>,
>> Steve Jaros <TrojanFamily...@trojannutcase.com> wrote:
>> >> Nobody says clutch hits don't exist. The existence of clutch hitters is
>> >what's
>> >> being debated.
>> >
>> >What would you consider to be evidence of a clutch hitter?
>>
>> Statistical significance.
Typically gentle but effective criticism from David follows below.
Statistical significance is not the same as 95% confidence level.
So consider my statement ammended. Something approaching 95% confidence
is required to start the discussion. This is after all the general
level of precision we require for a stat to be taken seriously.
>>
>> I'm prepared to argue that Paul Molitor and Tony Fernandez meet
>> a definition of clutch hitter (though I don't insist that they
>> are -- I merely note that they meet a particular test)
>>
>> Between 1984 and 1998 Paul Molitor hit .312 overall and .345 with
>> RISP. Sample size is just under 2,000 AB with RISP. Standard
>> deviation in this kind of sample size is around 9-10 points.
>
>For a .312 hitter, the variance of his batting average in 2000 AB with
>RISP is (.312)*(.688)/2000=.000107. The variance of his batting average
>in 6000 AB without RISP is (.312)*(.688)/6000=.0000036. Adding these
>variances and taking the square root gives a standard deviation of 12
>points, not 10 points.
I see the errors I made now that I've re-checked. First I worked from
emperical data (though I'd have expected that to produce a larger
error than theoretically predicted. But you've pointed out in the past
that this isn't the case). Second (for reasons of simplicity) I didn't
carry enough significant digits. A simplifying assumption compounded the
problem.
>The probability that a random player will exceed the expected value by
>at least 33 points, when the standard deviation is 12 points, is .003.
>In other words, if you looked at all 1998 regulars with long careers,
>and there were no ability, this is about where you would expect the top
>player to be, given that there are about 250 players. It is not
>statistically significant.
Right. As Perry noted you'd expect a couple of clutch/choke players even
if the ability didn't exist (at least using 95% confidence). And this is
a good practical example of why 95% confidence doesn't constitute proof.
Still as I said 95% is a reasonable place to start the discussion.
>Note that this does not disprove the existence of clutch hitting. If
>there is an ability to hit better with runners in scoring position, and
>the standard deviation of that ability is 5 points (one hit every other
>year), then the actual standard deviation over a career of Molitor's
>length is 13 points. But such an ability wouldn't mean much; the top 2%
>of clutch hitters would have an ability to get one extra hit a year.
Another way to look at this. Even if clutch hitting is an ability it
isn't very important. Best I can tell, it takes about 2,000 AB before
a single point of BARISP (increase that is) produces an extra run. IE 15
years of excellence for Molitor produced roughly 29 extra runs for his
teams. Under 2 runs a year from the very best means that the results
will be in the noise for most everybody else.
--
RNJ
Ron Johnson
>
>> I did a study of every player with 1,000+ PAs with RISP between 1984 and
>> 1998.
>>
>> They hit .275/.346/.432/.340 overall and .278/.369/.430/.345 with RISP.
>> (The last number being OBP with IBB removed. Didn't think to remove
>> SF and I don't feel like re-doing this. Won't change the essential
>> conclusion.)
>>
>> There is however only an 86% correlation between BA with RISP and
>> overall BA. Quite strong, but it's evidence that a little something
>> is probably going on with RISP.
>
>What would the correlation be if there were no ability, and you just
>picked a random sample of 1/4 of each hitter's AB as his RISP ones?
I picked a less than random ~1/3 sample (April/May) and got something
around 91%.
--
RNJ
David J. Grabiner
> In article <ur811x...@alumni.princeton.edu>,
> grab...@alumni.princeton.edu says...
> > Here's a study I did ten years ago in which I looked for this effect,
> > and found essentially no correlation, along with some possible reasons
> > that there should be a very slight correlation even if there is no
> > clutch ability. (For example, with the use of spot left-handed
> > relievers, good left-handed hitters with large platoon splits are less
> > likely to have the platoon advantage in the late innings of close
> > games.)
> >
> > http://remarque.org/~grabiner/fullclutch.txt
>
> Thanks very much. Just one question, though. When Voros did his DIPS
> research, he found little correlation on BA on balls in play by doing a
> similar year-to-year correlation study. Then Woolner(?) and Tippett did
> studies where they looked at larger periods of time, and seemed to find
> that a pitcher probably does control what happens on balls in play to a
> great extent than Voros may have thought, although still far less than
> our intuition may have led us to believe before Voros released his
> results.... Now, not to criticize your study in any way, but is there
> any possibility that the same thing can be true with your study?
Yes. My study, like any statistical study, has a margin of error. The
correlation between past and current clutch performance is .01, plus or
minus .07. That is, if the correct correlation is .15 (which would
imply an ability four times as strong as my estimate), there is still a
2% chance that a study the size of mine would compute a correlation of
.01.
In addition, my study cannot tell anything about the distribution. My
best estimate is that clutch ability has a standard deviation of 8
points of OPS. If this standard deviation is correct and the ability is
normally distributed, this means that 2% of all hitters are more than 16
points of OPS better than average in the clutch, and 2% are more than 16
points worse. However, it could be that most players have no ability,
but that 2% of players have ability around +40 and 2% have an ability
around -40. Statisticians usually assume that a distribution is normal
unless there is some reason that it should not be, or evidence that it
is not.
The non-normal distribution may have affected the DIPS study. It may be
the case that the same fraction of fastballs, curveballs, and sliders
turn into hits when they are in play, but knuckleballs are much less
likely to become hits. If this is the case, then knuckleballers would
allow lower batting averages than expected from their other statistics,
but there are so few knuckleballers that this would not create a large
overall correlation.
Kenny1111
grab...@alumni.princeton.edu says...
Thanks for the reply... I'll probably have to re-read some of your work
and Voros's work as well to set out all the ideas in my head, but this
helped clear some things up.